Skip to main content
8 events
when toggle format what by license comment
Feb 27, 2024 at 2:24 comment added mathdoge For those who don't like the Woodbury identity, you could do a simultaneous diagonalization to $X^T X$ and $x^T x$, because they are symmetric and PSD.
Dec 28, 2021 at 22:47 comment added AAL There was indeed a typo in the denominator: missing a factor of $(X'X)^{-1}$. I have edited the answer accordingly and checked that it fixes the counterexample. Thank you for pointing it out.
Dec 28, 2021 at 22:35 history edited AAL CC BY-SA 4.0
Corrected typo in equation
Dec 28, 2021 at 18:36 comment added whuber A careful reader, without enough rep to comment, has called to my attention an error in the application of this identity. Indeed, the denominator of the right hand side of the equality cannot be correct because it does not scale properly upon a simple rescaling of the variables. (Alternatively, consider $X=(1,1)^\prime$ (for $p=0$) and $x=(1).$ It asserts $(2)^{-1}=(2)^{-1} - (2)^{-1}(1)(2)^{-1}/(1+1),$ equivalent to $1/3=3/8.$) The argument is a good one, so it would be nice to get the details right. stats.stackexchange.com/a/90921/919 shows what the formula ought to look like.
Jun 24, 2020 at 16:28 review Late answers
Jun 24, 2020 at 17:03
Jun 24, 2020 at 16:22 comment added whuber +1. Thank you for the clear explanation -- and welcome to CV.
Jun 24, 2020 at 16:18 review First posts
Jun 24, 2020 at 18:38
Jun 24, 2020 at 16:11 history answered AAL CC BY-SA 4.0